Triple
T26056486
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Château de La Roche-Guyon |
E657130
|
entity |
| Predicate | hasEarlierStructure |
P159860
|
FINISHED |
| Object | medieval fortress |
—
|
LITERAL FINISHED |
How this triple was built (2 steps)
Every LLM step that produced this triple, in pipeline order — named-entity classification, the disambiguation choices (the exact options shown, with the pick highlighted), and the generated description. The batch + timestamp of each is in the Provenance table below.
NER
Named-entity recognition
gpt-5-mini
Instruction
Given a phrase, classify it is english named entity (e.g., persons, organizations, works of art) in Latin script, or not (e.g., literals, dates, URLs, verbose phrases). For disambiguation, the statement where the phrase occurs as object is also given. Please return a JSON object with `phrase` (string, the phrase being analyzed) and `is_ne` (boolean, indicating whether the phrase is a Named Entity).
Input
Phrase: medieval fortress | Statement: [Château de La Roche-Guyon, hasEarlierStructure, medieval fortress]
PD
Predicate disambiguation
gpt-5-mini-2025-08-07
Target predicate: hasEarlierStructure Context triple: [Château de La Roche-Guyon, hasEarlierStructure, medieval fortress]
-
A.
hasEarlierStructureOnSite
chosen
Indicates that a given site previously contained an earlier structure before the current one.
-
B.
hasEarlierVersion
Indicates that one entity is an earlier or prior version of another entity in a version sequence.
-
C.
hasPredecessorStructureFrom
Indicates that one structure existed or was established before another structure in a sequential or developmental order.
-
D.
hasEarlierPhase
Indicates that one phase occurs before another phase in a process or sequence.
-
E.
hasStructureAbove
Indicates that one entity has another entity positioned vertically higher or located on top of it within a structural or spatial arrangement.
- F. None of above.
Provenance (3 batches)
The batch behind each pipeline step, in order, with when it ran. Timestamps are batch-level — stages were processed in waves, so the object chain (NER → NED1 → NEDg → NED2) reads in order, but predicate / elicitation batches can sit in a different wave.
| Step | Stage | Batch ID | Status | When |
|---|---|---|---|---|
| creating | Elicitation | batch_69ee5bbd788481909e22bd7153d0c037 |
completed | April 26, 2026, 6:38 p.m. |
| NER | Named-entity recognition | batch_69f6068e8b188190b445063c1c03dfb8 |
completed | May 2, 2026, 2:13 p.m. |
| PD | Predicate disambiguation | batch_69f602d07590819085ac34b189613104 |
completed | May 2, 2026, 1:57 p.m. |
Created at: April 26, 2026, 7:10 p.m.